The origins of qualia 1

Transcription

1 From Tim Crane & Sarah Patterson (eds.) The History of the Mind-Body Problem (London: Routledge 2000) The origins of qualia 1 Tim Crane 1. The contemporary mind-body problem The mind-body problem in contemporary philosophy has two parts: the problem of mental causation and the problem of consciousness. These two parts are not unrelated; in fact, it can be helpful to see them as two horns of a dilemma. On the one hand, the causal interaction between mental and physical phenomena seems to require that all causally efficacious mental phenomena are physical; but on the other hand, the phenomenon of consciousness seems to entail that not all mental phenomena are physical. 2 One may avoid this dilemma by adopting an epiphenomenalist view of consciousness, of course; but there is little independent reason for believing such a view. Rejecting epiphenomenalism, then, leaves contemporary philosophers with their problem: mental causation inclines them towards physicalism, while consciousness inclines them towards dualism. To accept that this is the way that the problem is generally conceived is not to accept that the problem has been well-formulated. One may legitimately question the assumptions which give rise to the mental causation problem: for instance, what are 1 This paper was originally written for a conference on the future of analytic philosophy at the University of Warwick in Later versions were presented at a conference in Miskolc, Hungary, and at the Universities of Cambridge, Uppsala, Helsinki and University College Cork. Thanks to Katalin Farkas, Peter Lipton, Greg McCulloch, Howard Robinson and Jonathan Wolff for comments. I would like to express a special debt to Mike Martin for invaluable criticism and the influence of his writings. 2 For this way of seeing the mind-body problem as a dilemma, see Tim Crane The mind-body problem in F. Keil and R. Wilson (eds.) MIT Encylopedia of Cognitive Science (Cambridge, Mass.: MIT Press 1999). See also Jaegwon Kim, The mind-body problem after fifty years in A. O Hear (ed.) Current Issues in the Philosophy of Mind (Cambridge: Cambridge University Press 1998) p.21, and David Braddon-Mitchell and Frank Jackson, Philosophy of Mind and Cognition (Oxford: Blackwell 1996) pp

2 the grounds for believing that anything that interacts with something physical must itself be physical? And how can we formulate a conception of the physical which does not render the question trivial (as would be done by treating the physical as the causal)? 3 However, as an account of the current state of the debate within contemporary philosophy of mind, the above description of the problem ought to be fairly uncontroversial. It is perhaps equally uncontroversial to say that while the problem of mental causation is regarded as a relatively technical problem whose solution requires only a more careful treatment of the notions of causation or physical realisation the problem of consciousness is thought to be a deeper and more difficult problem for physicalists and non-physicalists alike. The problem is often expressed in terms of qualia : the qualitative or phenomenal features of conscious states of mind. How can a mere physical object, which we know a person to be, have states of mind with qualitative features or qualia? This question, which poses what David Chalmers calls the hard problem of consciousness, is supposed to be the real heart of the mind-body problem in today s philosophy. 4 As Jaegwon Kim puts it, the stance you take on the problem of qualia [is] a decisive choice point with respect to the mind-body problem. 5 To have a clear understanding of this problem, we have to have a clear understanding of the notion of qualia. But despite the centrality of this notion in formulating this aspect of the mind-body problem, it seems to me that there is not a clear consensus about how the term qualia should be understood, and to this extent the contemporary problem of consciousness is not well-posed. The difficulty here can be vividly brought out at first by considering the fact that there seems to be a real dispute about whether qualia exist at all. Anyone with 3 See Tim Crane, The mental causation debate Proceedings of the Aristotelian Society Supplementary Volume 95, 1995, and Against physicalism in S. Guttenplan (ed.) A Companion to the Philosophy of Mind (Oxford: Blackwell 1994). 4 See David Chalmers, The Conscious Mind (Oxford: Oxford University Press 1996). 5 Kim, The mind-body problem after fifty years p

3 the slightest familiarity with the recent debate will be aware that some philosophers take the existence of qualia to be an obvious fact, while others deny their existence. So on the one hand, we find Ned Block responding to the question, what is it that philosophers have called qualitative states? with the quip: As Louis Armstrong said when asked what jazz is, If you got to ask, you ain t never gonna get to know. 6 But on the other hand, we find Michael Tye and Gilbert Harman arguing that there are no qualia in visual experience, and Daniel Dennett denying the existence of any qualia whatsoever. In Consciousness Explained, Dennett says: Philosophers have adopted various names for the things in the beholder (or properties of the beholder) that have been supposed to provide a safe home for the colors and the rest of the properties that have been banished from the external world by the triumphs of physics: raw feels, sensa phenomenal qualities, intrinsic properties of conscious experiences, the qualitative content of mental states and of course qualia...there are subtle differences in how these terms have been defined, but... I am denying there are any such properties. 7 At first sight, this dispute might seem to be a straightforward ontological matter, like a dispute about the existence of numbers or universals. But closer reflection shows that the dispute cannot be exactly like this. For the normal route to introducing numbers or universals into an ontology is that they explain some phenomenon which is agreed on all sides to exist and require an explanation: mathematical practice, or apparent sameness of kind. The claim is that we should believe in these entities because they explain the obvious truths about the appearances, broadly understood. But the truths about qualia, by contrast, are supposed to be truths about the appearances themselves, about how things seem to us in experience. And it is reasonable to expect that how things seem to us should not be a theoretical posit, but a pre-theoretical starting point: 6 Ned Block, Troubles with functionalism in Block (ed.) Readings in the Philosophy of Psychology, Volume I (London: Methuen 1980) p Consciousness Explained (London: Allen Lane 1991) p.372 3

4 a point from which to embark on a debate, where things are relatively obvious to all its participants. This is clear from Block s remark: that there are qualia is supposed to be something obvious, something whose existence a moment s reflection on experience or consciousness is supposed to establish. A theory of consciousness is a theory of qualia: qualia are the data to be explained by a theory of consciousness. A similar approach is taken by Kim and Chalmers. Yet Dennett, Tye and Harman take themselves to be giving theories of experience while denying the existence of qualia (or at least of qualia of a certain kind). And Fred Dretske and W.G. Lycan argue, against Block, that qualia should be conceived as intentional states. 8 Furthermore, Tye and Harman argue against qualia by reflecting on experience, and claiming that it is obvious that there are no such things as qualia in perceptual experience. What Block claims obviously exist Tye and Harman claim obviously don t exist. What is going on? How can there be such extreme disagreement about what is obvious? 9 The aim of this paper is to gain an understanding this dispute by examining some of its historical origins in Twentieth Century philosophy. In particular, I shall be concerned with the origins of the terms quale and qualia, and with how these terms became central in the formulation of the mind-body problem. I shall claim that the debate will not be advanced by focussing inwards (to use a phrase of Colin McGinn s) and swapping intuitions about what is obvious. Rather, we should try and make progress by understanding how the term qualia (the concept of qualia) has entered the debate, by trying to see what role the term was introduced to play and for what purpose, and how this purpose may or may not differ from the purposes of contemporary disputants. The question on which I want to focus is: what was the 8 See Fred Dretske, Naturalizing the Mind (Cambridge, Mass.: MIT Press 1995); W.G. Lycan, Consciousness and Experience (Cambridge, Mass.: MIT Press 1996). 9 My understanding of this issue is indebted to M.G.F. Martin, The transparency of experience, forthcoming. This paper, and the opening chapters of Martin s forthcoming book, Uncovering Appearances, give the best account of (and the most convincing solution to) the problem of the disagreement about the obvious in the theory of perception. 4

5 original point of talking in terms of qualia, and does this point differ from the point which contemporary philosophers are trying to make by using the term? I begin by examining a connection between the dispute about qualia and a dispute which many would now consider defunct: the debate over sense-data in the philosophy of perception. I argue that a similar puzzle arises there as arises over qualia, and that the solution to the puzzle gives the resources to suggest, in broad outline, how the dispute over qualia may be resolved. Section 3 then gives an account of the origin of the notion of qualia, and its relation to the notion of sense-data. I claim that in the philosophy of mind of the first half of this century, qualia and sense-data play similar roles: sense-data are objects of experience and qualia their properties. In section 4, I explain how qualia came to play a different role in subsequent discussions of the mind-body problem: rather than being properties of the objects of experience, they became properties of the experiences themselves. Section 5 outlines why this change took place, and suggests the conclusion that the most plausible conception of qualia in contemporary philosophy is an intentionalist one (or, better, a conception which rejects qualia as they are normally understood). This final conclusion is, strictly speaking, independent of the investigation into the origins of qualia but I nonetheless believe that the proper lesson of the historical investigation is that to appreciate the origins of qualia is to appreciate why we no longer need to talk in terms of them. 2. The origins of sense-data The idea that perceptual experience involves awareness of sense-data was a dominant theme in discussions of perception and epistemology in the first half of this century. 10 In contemporary discussion, by contrast, little attention is paid to the idea of sensedata, or to the question to which they were supposed to provide an answer, what are 10 A good anthology of key essays on the subject of sense-data is Perceiving, Sensing and Knowing ed. R. Swartz (Los Angeles and Berkeley: University of California Press 1965). 5

6 the direct objects of perception?. 11 But it has not been sufficiently noted that a puzzle emerged about sense-data that is closely parallel to the puzzle noted above about qualia. In a paper published in 1936, called Is there a problem about sense-data?, G.A. Paul remarked that Some people have claimed that they are unable to find such an object [as a sense-datum] and others have claimed that they do not understand how the existence of such an object can be doubted. 12 Sense-data were supposed to be the immediate objects of experience. So conceived, our awareness of sense-data are among those facts about experience which are open to philosophical reflection, rather than scientific theorising. One would expect, then, that awareness of sense-data is something that can be gleaned from thinking about what it is like to have an experience, in Nagel s phrase. Since facts about what it is like to have experience ought, on the face of it, be obvious to us, or obvious on reflection, then the existence of sense-data ought to be obvious to us. If this is so, then how can the situation described by Paul arise? How can it be that some philosophers deny that they find sense-data in their experiences, and others claim that it is obvious it cannot be denied that there are sense-data? This is the puzzle, parallel to our puzzle about qualia, towards which Paul s remark points. To solve this puzzle, we need to know more about the role of the concept of sense-data in the philosophy of the early 20th century. Many students of philosophy first encounter the sense-datum theory of perception in the opening chapters of Russell s 1912 The Problems of Philosophy. Like much in Russell s philosophy of this period, the doctrines advanced in these chapters reflect strongly the influence of G.E. 11 For an account of this change of emphasis in 20th century philosophy of perception, and of the underlying, persisting problem of perception, see Martin, A problem of interpretation, this volume. 12 G. A. Paul, Is there a problem about sense-data? (1936) reprinted in Swartz (ed.) Perceiving, Sensing and Knowing p

7 Moore. 13 And it was Moore who was the first to introduce the term sense-data into philosophy, in his lectures called Some Main Problems of Philosophy. 14 But there is a difference in the way that Moore thought of sense-data, and the way they have come to be conceived in the later philosophical discussion. In contemporary discussions, the sense-datum theory of perception is often put forward only to be refuted by the manifest absurdity of its commitment to mysterious non-physical objects, the sensedata of which we are supposedly aware. 15 This commitment was influentially criticised by J.L. Austin, 16 and would be rejected by many naturalistic philosophers today on broadly metaphysical grounds. Now certainly, the idea of being aware of (and therefore causally affected by) non-physical objects seems to be suggested by Russell s account in the Problems. But in his initial discussions, Moore did not introduce the term as having this meaning. In his lectures, Moore defines the term sense-data by using an example of looking at a white envelope. He then claims that what are seen are patches of colour and shapes: These things: this patch of a whitish colour, and its size and shape I did actually see. And I propose to call these things, the colour and size and shape, sense-data, things given or presented to the senses. 17 Moore commented much later (1952) that he should have called the patch the sensedatum, and not the properties of the patch the colours are properties of the particular patch. In another lecture in the series, Moore employs a similar strategy. In 13 See The Problems of Philosophy (Oxford: Oxford University Press 1912), where Russell explicitly acknowledges Moore s influence on his views on perception; Peter Hylton, Russell, Idealism and the Emergence of Analytical Philosophy (Oxford: Oxford University Press 1989), and Thomas Baldwin, G.E. Moore (London: Routledge 1990), e.g., p.323 fn Moore s most important writings on perception are collected in T. Baldwin (ed.) G.E. Moore: Selected Writings (London: Routledge 1993) See also Baldwin, G.E. Moore, pp See, for example, W.H.F. Barnes, The myth of sense-data ( ) in Swartz (ed.) Perceiving, Sensing and Knowing 16 Sense and Sensibilia (Oxford: Oxford University Press 1962) 17 Moore, Selected Writings p.48. XXX 7

8 asking us to look at a pencil, he says that what you directly apprehend is a patch of brownish colour, bounded on two sides by fairly long parallel straight lines. These coloured patches are the sense-data you see. 18 Moore distinguishes between the sense-datum, which is given to the mind, and the sensation, which is the act or event of being aware of the datum. (This is what used to be called an act-object conception of experience.) Having made this distinction, we can see that two possibilities emerge, within the terms of the definition put forward so far. One is that sense-data are mind-independent objects presented in experience. The other is that sense-data are not mind-independent objects. Are sense-data objects of the first kind, or of the second? Moore s answer to this question oscillates, throughout all his discussions of perception. In his lectures, he denies that sense-data are ordinary material objects, since while two people can see the same object, no two people can sense the same sense-datum. 19 This suggests an indirect realist account of perception: sense-data, non-physical objects, are the immediate objects of experience in virtue of which we are aware of physical objects. But, in a paper, Some judgements of perception, we find him reverting to a direct realism (at least about the objects of perception if not how they appear). Perhaps because of this oscillation, Moore s first attempts to define sense-data did not meet with universal acceptance. 20 So in a later paper, A defence of common sense (1925) he attempts to define them again in terms that should be so uncontroversial that there is no doubt at all that there are sense-data, in the sense in which I am now using that term. 21 Again, his method is ostensive: In order to point out to the reader what sort of things I mean by sense-data, I need only ask him to look at his own right hand. If he does this he will be able to pick out something... with regard to which he will see that... it is a natural 18 Hume s theory examined in Selected Writings, p Selected Writings, ed. Baldwin p.51. XXX 20 See G.F. Stout, XXX 21 A defence of common sense in Selected Writings, ed. Baldwin p

9 view to take that that thing is identical, not, indeed, with his whole right hand, but with that part of his surface which he is actually seeing, but will also (on a little reflection) be able to see that it is doubtful whether it can be identical with the part of the surface of his hand in question. Things of the sort (in a certain respect) of which this thing is, which he sees in looking at his hand, and with regard to which he can understand how some philosophers should have supposed it to be the part of the surface of his hand which he is seeing, while others have supposed that it can t be, are what I mean by sense-data. 22 Moore adds that he meant to define the term in such a way that it should be an open question whether or not the sense-datum is identical with part of the surface of his hand. This attempt to clarify the position was famously criticised by O.K. Bouwsma on the grounds that the procedure cannot identify sense-data in this neutral way unless we already have Moore s conception of what sense-data are. 23 If we follow Moore s instructions and pick out the surface of our right hand, we will not be able to doubt whether the thing we have picked out is the surface of our right hand unless we have already picked it out as something which might not be the surface of our right hand: that is, as a sense-datum. It is Moore s conception of sense-data which is driving the possibility of doubt, not vice versa. However, Bowsma misses Moore s point. What Moore is trying to do in this passage, as the last quoted makes plain, is to bring out the sense in which (almost) all philosophers have agreed on something. They have disagreed about whether one sees the surface of one s hand: some say that what is seen is the surface of one s hand, others deny that it is the surface of one s hand. But what is the it I am talking about when I say that others deny that it is the surface of one s hand? The it is what Moore means by sense-data: the object of experience, whatever it is. All one needs to do to understand this definition of Moore s is to understand how a philosopher might doubt that what one sees when one seems to see the surface of one s hand is indeed the 22 A defence of common sense pp O.K. Bouwsma, Moore s theory of sense-data in P. Schilpp, The Philosophy of G.E. Moore (La Salle, Illinois: Open Court 1942). 9

10 surface of one s hand. To do this is not to propose any particular philosophical theory about the object: it is just to entertain the idea of an object of experience. The matter becomes clearer when we consider H.H. Price s views in Perception (1932). Price, who had attended Moore s lectures in Cambridge, introduces the notion of sense-data in Moore s way: The term sense-datum is meant to be a neutral term... The term is meant to stand for something whose existence is indubitable (however fleeting) something from which all theories of perception ought to start. 24 To illustrate what kind of things these are whose existence is indubitable, Price introduced the example of seeing a tomato: When I see a tomato there is much that I can doubt. I can doubt whether it is a tomato I am seeing, and not a cleverly painted piece of wax. I can doubt whether there is a material thing there at all... One thing however I cannot doubt: that there exists a red patch of a round and somewhat bulgy shape, standing out from a background of other colour-patches, naad having a certain visual depth, and that this whole field of colour is presented to my consciousness... that something is red and round then and there I cannot doubt... that it now exists and that I am conscious of it by me at least who am conscious of it this cannot possibly be doubted... This peculiar and ultimate manner of being present to consciousness is called being given, and that which is thus present is called a datum. 25 Price points out here that what can be doubted here is that it is a tomato I am seeing, or that the thing is a material object. But what cannot be doubted is that there is something red and round which I am seeing. Like Moore, then, Price introduces his reader to the idea of sense-data as the entities, whatever they are, which are present to consciousness in experience in this peculiar and ultimate manner. Since J.L. Austin s criticism of the sense-datum theory in Sense and Sensibilia (1962) it has become common to attribute the errors of the sense-datum theory to 24 Perception (London: Routledge and Kegan Paul, 1932) p Perception p.3. 10

11 foundationalist, infallibilist epistemology: sense-datum theorists were trying to find the certain basis of all knowledge. 26 And the conclusion is drawn from this that once we dispense with these epistemological requirements, the motivation to posit sense-data evaporates. But, as M.G.F. Martin has made clear, the natural interpretation of the passage from Price does not require attributing to him these epistemological motives. 27 Price s point is essentially the same as Moore s: that perception presents us with something, something is given to us this something is the given. Price distinguishes two methods by which we may distinguish different senses of the given, the physiological and the immanent or phenomenological. The physiological enquiry into perception delivers the given in a causal sense: the immediate causal precursors of experience. The immanent method delivers the given in the phenomenological sense: what is given to consciousness or presented to the mind. Price s point, then, is that perceptual experience is relational: experience relates us to something which is given to us. 28 Among recent writers, Howard Robinson endorses this view in what he calls his Phenomenal Principle : that when one has an sensory experience as of something being F, there is something F which one is experiencing. 29 Now this principle may be controversial and we shall find reasons for doubting it but it is not controversial because it entails a foundationalist infallibilist epistemology, or an attempt to refute scepticism. And if we compare Robinson s Phenomenal Principle with Moore s and Price s remarks about sense-data, it is plain that Robinson is essentially following through the same lines of thought as they were. 30 C.D Broad argued along the same lines that when we approach perception 26 Austin s diagnosis of the errors of sense-data theorists was that their real motive is that they wish to produce a species of statement which is incorrigible (Sense and Sensibilia p.103). 27 See Martin, J.L. Austin: Sense and Sensibilia revisited forthcoming. 28 Perception, p Perception (London: Routledge 1994) chapter For other writers who argue for sense-data without making epistemological assumptions, see F. Jackson, Perception (Cambridge: Cambridge University Press 1977); E.J. Lowe, Experience and its objects in Tim Crane (ed.) The Contents of Experience (Cambridge: Cambridge University Press 1992). 11

12 from what he calls the purely phenomenological point of view perception is ostensibly prehensive of the surfaces of distant bodies as coloured and extended. 31 Now Price thought it a gross absurdity to suppose that the existence of sensedata depend on our awareness of them. 32 But nonetheless, he could not straightforwardly identify sense-data with material objects or their surfaces. The reason comes from the essential consideration underlying the argument from illusion: if my experience could be the same even if the material object of the experience did not exist, then the material object is not essential to the experience. But since it is indubitable that experience is relational, that all experiences have objects, then the objects of experience are not material objects. And since what is present to our minds does not exist in our minds, then we are led to the conclusion that sense-data are nonmental, non-physical objects. The argument involves many steps and many assumptions, and it is not my aim to discuss it in detail here. The important point at this stage is that the argument which leads sense-data theorists to the idea that sensedata are non-physical objects is a further step, further to the introduction of the idea of sense-data itself. This brief survey has given us enough material to solve the puzzle in Paul s article. What Moore and Price think is obvious is that phenomenologically, perception has a relational character: that experience involves being given something. To this extent, they think that the existence of sense-data cannot be denied. But a more demanding conception of sense-data arises from reflection on how to reconcile the idea that experience is relational with the apparent possibility that a perceiver could be in an indistinguishable mental state from a perception of a mind-independent object, 31 C.D. Broad, Elementary reflexions on sense-perception (1952) in Swartz (ed.) Perceiving, Sensing and Knowing, p.32. By the phenomenological point of view, Broad means as they appear to any unsophisticated perceipient, and as they inevitably go on appearing even to sophisticated percepients whose knowledge of the physical and physiological processes assures them that the appearances are misleading p. 30. By prehension Broad means what Russell meant by acquaintance. 32 Perception p

13 and yet the mind-independent object not exist. If we allow this possibility, and we hold that experience is relational, then it is a short step to admitting that what is given cannot be a mind-independent object. Since the experience is qualitatively identical in both cases, the conclusion is drawn that the object of the experience is the same in both cases assuming, of course, that the nature of the experience is supervenes on the nature of its objects. And it is this conception of sense-data as objects which the opponents of sense-data claim not to be able to find in experience. In short, when Moore and Price say that the existence of sense-data cannot be denied, what they mean is that it cannot be denied that experience is relational. But when the critics of sense-data say they cannot find sense-data in experience, they are questioning the existence of sense-data in the more demanding sense, the sense which is generated by the conception of experience as relational plus the argument from illusion. This, it seems to me, is the solution to the puzzle posed by G.A. Paul s remarks. 33 In fact, it can be denied that experience is relational. This is, after all, what an adverbial theory of perception says. An adverbialist holds the qualities sensed in experience to be modifications of experience itself: sensing a blue patch ought to be understood as sensing bluely. This theory does deny what Moore and Price take to be obvious. But it is not hard to understand how such a theory could have arisen out of a resistance to sense-data in the demanding sense: since it is absurd to suppose that there are such strange objects outside the mind, the distinctions in experience which these objects are supposed to mark must really be modifications of the experiences themselves Cf Barnes, The myth of sense-data (1944-5) in Perceiving, Sensing and Knowing p.159. See also chapters 1 and 2 of J.J. Valberg, The Puzzle of Experience (Oxford: Oxford University Press 1992). 34 See C. Ducasse s criticism of Moore s theory, and defence of adverbialism, in Moore s Refutation of Idealism in Schilpp (ed.) The Philosophy of G.E. Moore. 13

14 As we shall see, a similar distinction between those who think that differences between experiences are explained in terms of differences in their objects, and those who think that some such differences are explained in terms of differences in properties of the experiences will be what provides the key to the puzzle about qualia posed in section 1. This should not be surprising if I am right in my claim that sense-data and qualia were introduced to play very similar roles. I shall attempt to argue for this claim in the next section. 3. The origins of qualia While sense-data are largely a British invention, it is American philosophy which can lay claim to the invention of qualia. The first philosopher to use terms quale and qualia in something like its modern sense was C.S. Peirce. When Peirce wrote in 1866 that there is a distinctive quale to every combination of sensation... a peculiar quale to every day and every week a peculiar quale to my whole personal consciousness, he was talking about what experience is like, in a general sense, not restricted to the qualia of experience in the sense in which it is normally meant today. 35 William James occasionally used the term specifically to discuss sensation, but as far as I can see the term had no special technical significance in his philosophy or psychology. 36 The term is occasionally used by some of the so-called New Realists: for instance, we find R.B. Perry talking of sensory qualia which are localisable in the body but as with James the term means little more than sensation C.S. Peirce, Collected Papers para 223. (PLACE PUBLISHER) For today s usage, see the discussion in section 4 and 5 below. 36 See for example chapter 20 of his Principles of Psychology, the on The meaning of localization : No single quale of sensation can, by itself, amount to a consciousness of position... a feeling of place cannot possibly form an immanent element in any single isolated sensation. In The Origins of Pragmatism (San Fransisco: Freeman 1968) A.J. Ayer puts forward a theory of experience in terms of qualia in the context of a discussion of James; but it is clear that Ayer does not attribute this conception of qualia to James. For Ayer, qualia are essentially sense-data in the sense discussed in section 2 above: see pp.299 and R.B. Perry, TITLE in Journal of Philosophy 9, 1914, p.153; see also the discussion in J. Dewey, Value, objective reference and criticism Philosophical Review VOLUME 14

15 It is fairly clear that the chief source of the technical use of the term qualia is C.I. Lewis s discussion in Mind and the World Order (1929). 38 The theme of this work is to reconcile what Lewis saw to be correct in the idealism of those such as Royce, who held that experience always involves interpretation or conceptualisation, and the realism of his day, which held (along with Russell and Moore) that whatever the mind grasps must be independent of the mind. The basis of Lewis s reconciliation is a distinction between two elements in our cognitive lives: the immediate data which are presented or given to the mind and the construction or interpretation which the mind brings to those data. 39 Lewis rejects the idealist critique that any experience is so entirely conceptualised that there is no non-conceptualised core: The distinction between this element of interpretation and the given is emphasised by the fact that the latter is what remains unaltered, no matter what our interests, no matter what we think or conceive. (p.52) But the fact that there is this unaltered given does not entail that we are able to describe it, since in describing it... we qualify it by bringing it under some category or other, select from it, emphasise aspects of it, and relate it in particular and unavoidable ways (p.52). So the given is in a sense ineffable: The absolutely given is a specious present, fading into the past and growing into the future with no genuine boundaries. The breaking of this up into the presentation of things marks already the activity of an interested mind. (p.58) Having made this claim, however, Lewis then goes on to say a little more about what it is that is given: 38 Mind and the World Order (London: Constable 1929). It is not difficult to see where Lewis might have got the term from: his teachers at Harvard as an undergraduate were James and Royce, and his postgraduate teacher in was Perry. 39 Mind and the World Order p

16 In any presentation, this content is either a specific quale (such as the immediacy of redness or loudness) or something analyzable into a complex of such. The presentation as an event is, of course, unique, but the qualia which make it up are not. They are recognisable from one to another experience. (p.60) Qualia, then, are properties of what is given. However, although they are universals, since they can re-occur in distinct experiences (p.121), qualia should not be confused with the objective properties of objects in the external world. The critical realists R.W. Sellars, C.A. Strong, together with C.D. Broad are criticised for making this confusion. Lewis s reasons for distinguishing qualia from objective properties are that objective properties are always more complex in nature than qualia, and their existence extends beyond the specious present. The same objective property, for instance blueness, can give rise to many different colour qualia in different situations (p.121). It is the confusion between properties of objects and qualia that gives rise, in Lewis s view, to the absurdity of the idea of unsensed sensa a problematic idea with which those in the sense-data tradition struggled. Objective properties are what we have knowledge of; we have no knowledge of qualia since knowledge always transcends the immediately given (p.132). How then does Lewis allow himself to talk of the immediacy of redness or loudness as qualia? His answer reveals a commitment to the possibility of inverted qualia: Qualia are subjective; they have no names in ordinary discourse but are indicated by some circumlocution such as looks like... All that can be done to designate a quale is, so to speak, to locate it in experience, that is, to designate the conditions of its recurrence or other relations of it. Such location does not touch the quale itself; if one such could be lifted out of the network of its relations, in the total experience of the individual, and replaced by another, no social interest or interest of action would be affected by such substitution. (p.124) 16

17 We can talk about qualia, then, by comparing them to properties in the world and by locating them in terms of their relations to what brings them about. But our access to qualia in this way will be inevitably indirect. Qualia are properties, then; but what are they properties of? They cannot be properties of material objects for the reason stated. Lewis s answer is they are properties of an event, the event which is the presentation of the given (p.60). Qualia are the recognisable qualitative characters of the given (p.121). Lewis actually says that what is given may exist outside the mind that question should not be prejudiced (p.64). Initially, it is hard to square this thought with the idea that qualia are subjective (p.124). However, Lewis s remarks could be defended on the grounds that exists outside the mind does not entail mind-independent. An object may be minddependent even though it exists outside the mind: it may be an object of awareness, distinct from the state of mind which is the awareness of it, which is brought into existence by the state of awareness itself. This recalls the conception of sense-data as the given: non-physical, non-mental items which are present to the mind in experience. So what actually is the difference between qualia and sense-data? As we have noticed, one difference is simply a difference in metaphysical category: qualia are properties (so, universals) and sensedata are particulars. But this need not be significant, since sense-data have properties too, and qualia are the properties of a particular, the given. What exactly would be wrong with an interpretation of qualia- and sense-datum theories which treats sensedatum as the given, and qualia as the properties of the given? Many of Lewis s contemporaries and commentators saw the matter that way. E.M. Adams explicitly identifies sense-data and qualia (though, as we have seen, this is to confuse particulars and universals) and Lewis s student Roderick Firth 17

18 straightforwardly asserts that Lewis was a sense-datum theorist. 40 This interpretation is to true to the common meaning of datum and given : after all, and Lewis clearly would reject an adverbialist conception of experience. Both Lewis and the sense-data theorists emphasise the given element in experience, and like Price, Lewis finds it to be a plain fact about conscious experience that something is given in experience. As he says, no-one but a philosopher could for a moment deny this immediate presence in consciousness of that which no activity of thought can create or alter (p.53). 41 However, the identification of qualia with the properties of sense-data would be resisted by many contemporary philosophers who are happy to accept qualia (in their sense) but will have nothing to do with sense-data, or even the given in any form. Many of these philosophers would agree with the view Firth attributes to Lewis: Lewis never makes the mistake that Thomas Reid so eloquently charged to Descartes and the British Empiricists the mistake of treating sense-experience as the object of perception. 42 The point attributed here to Reid became a common criticism of sense-data theories: we do have sensations in perception, the objection runs, but that doesn t mean we perceive those sensations. Rather we perceive objects by having sensations that is the only role of sensation in perception. Now it is true that Lewis will not say that we 40 E.M. Adams, C.I. Lewis and the inconsistent triad of modern empiricism pp.377 and 384; Roderick Firth, Lewis on the given p.330, both in The Philosophy of C.I. Lewis ed. P.A. Schillp (La Salle Illinois; Open Court 1968). 41 Further reason for identifying qualia with the properties of sense-data can be derived by looking at Goodman s use of the concept of qualia in The Structure of Appearance (Cambridge, Mass: Harvard University Press 1951). For Goodman, qualia are the primitives of a phenomenalist system of accounting for the whole of reality: they are the phenomenal individuals out of which enduring public objects are constructed. To this extent, they resemble the sense-data of Ayer, in The Central Questions of Philosophy (London: Weidenfeld 1973). One thing that makes this discussion difficult to integrate with the contemporary issues about qualia is that the distinction between the physicalistic and the phenomenalistic systems is supposed to be a difference in choice of language rather than in any matter of fact. On this aspect of Ayer s view, see Martin, Austin: Sense and Sensibilia Revisited. 42 Firth, Lewis on the given p

19 perceive experiences, since an experience is the state of being aware of the given. But we are aware of the given: the given is not a sensation, but this is not a reason for saying it is not a sense-datum. The criticism of sense-data theories just mentioned misses the point, since as we saw with Moore and Price, sense-datum theories are careful to distinguish the sense-datum from the act of sensing it. They are not committed to the view that we see sensations. The heart of the matter, it seems to me, is this. Lewis s qualia-theory and the sense-datum theory resemble each other in their central claim: that in experience, something is given. They differ in that Lewis thinks that the qualitative properties of experience are in a sense ineffable, and can only be indirectly described; but in the context of their common commitment to the given, this is a relatively unimportant disagreement. As far as the core commitments of the two theories go, it would not mislead to say that the given is a sense-datum, and qualia are its properties. There is a radical difference in contemporary philosophers attitude to qualia and their attitude to sense-data. Contemporary philosophers are fairly unanimous in their rejection of sense-data. The idea that experience is not awareness of non-physical objects is thought to be an out-dated product of a discredited epistemology and philosophy of mind. But it is perhaps equally clear that there are as many contemporary philosophers who accept the existence of qualia as there are those who reject sense-data. Sense-data are the product of confusion; qualia, on the other hand, are troublesome but undeniable features of our experience of which we have to give a physicalist or naturalist account. What has all this got to do with the mind-body problem? Lewis and Goodman both had theories in which qualia play an important role; but they did not see this as relevant to the question of the relationship between mind and body. We can trace the explicit interest in the bearing of qualia on this relationship to Herbert Feigl s

20 essay, The Mental and the Physical. 43 Feigl accepts the very persuasive arguments [which] point simply to the existence... of immediate experience, i.e. the raw feels or hard data of the immediately given. 44 But he explicitly distinguishes his own form of empiricism from C.I. Lewis s epistemology, according to which physical knowledge concerns only the form or structure of events in the universe, whereas acquaintance concerns the contents or qualia of existence. 45 The difference is that whereas Lewis denied that qualia can be known at all, Feigl thought that they could be known by acquaintance (for Lewis, acquaintance is not a form of knowledge). The given element in immediate experience gives rise to the mind-body problem: for having allowed that there can be knowledge of qualia, Feigl explicitly states what has now become known as the knowledge argument (though he does not, of course, draw the conclusion that physicalism is false). This gives us one link between Lewis s conception of qualia and the current problem of qualia for physicalism. But given the many ways that the term qualia has been used, do we have any firmer analytic reasons to suppose that the qualia which pose the problem for physicalism are the qualia which are the properties of sense-data or the given? I think there is, but teasing out the connection is a little complex: this is the task of the next section. 4. The contemporary problem of qualia When qualia are discussed in contemporary philosophy, how are they being conceived? As might be expected, there is not one answer to this question, since in the 43 The Mental and the Physical first published in Minnesota Studies in the Philosophy of Science eds. H. Feigl, M. Scriven and G. Maxwell (Minneapolis: University of Minnesota Press 1958). Reprinted with a postscript by the same publisher in Lawrence Nemirow makes the explicit connection between his presentation of the problem and Feigl s in his Physicalism and the cognitive role of acquaintance in W.G. Lycan (ed.) Mind and Cognition (Oxford: Blackwell 1987). 44 The mental and the physical p.8. The expression raw feel is from E.C. Tolman, Purposive Behaviour in Man and Animals (PLACE PUBLISHER 1932). 45 The mental and the physical p

21 contemporary debate, the term qualia is used in a number of different ways. Dennett says qualia are supposed to be properties of a subject s mental states that are 1. ineffable; 2. intrinsic; 3. private; 4. directly or immediately apprehensible in experience 46 Sydney Shoemaker describes the objection that functionalism cannot account for qualia as substantially the same as the objection that functionalism cannot account for the raw feel component of mental states, or for their internal or phenomenological character. 47 And Scott Sturgeon has said that qualia are just those properties in virtue of which the experience has the conscious subjective character it has. 48 Frank Jackson and David Braddon-Mitchell distinguish the use of the word qualia to denote non-physical properties of the mind (according to which the existence of qualia is incompatible with physicalism) and the use of the word to denote intrinsic non-functional properties (according to which the existence of qualia is incompatible with functionalism). 49 And M.G.F. Martin distinguishes (in the contemporary debate) between qualia as conceived as properties of experiences, and qualia conceived as apparent properties of the objects of the experiences. 50 What should we make of these very different uses of the term? The first thing to notice is that the variety of uses provides us with a straightforward way of solving the puzzle of section 1, in a parallel way to the solution provided with sense-data. As with sense-data, there is a relatively innocuous sense of qualia (roughly, Sturgeon s) where for a state to have qualia is just for it to be a conscious state. In this sense, Block is right that the existence of qualia cannot be denied. But if qualia are the things Dennett and Jackson/Braddon-Mitchell are talking about, qualia in the more 46 D. Dennett, Quining qualia in W.G. Lycan (ed.) Mind and Cognition p S.Shoemaker, Functionalism and qualia in Identity, Cause and Mind (Cambridge: Cambridge University Press 1980) p The epistemic view of subjectivity Journal of Philosophy DATE VOLUME 49 David Braddon-Mitchell and Frank Jackson, Philosophy of Mind and Cognition (Oxford: Blackwell 1996) pp See M.G.F. Martin, Setting things before the mind in Current Issues in the Philosophy of Mind ed. A. O Hear (Cambridge: Cambridge University Press 1998) pp

22 demanding sense, then it is not obvious that there are qualia. This predictable solution to our puzzle parallels the solution for sense-data. Solving this puzzle, however, does not resolve the substantial debate. For we still need to know what qualia in the more demanding sense are, and whether there are any such things. Here, since the mind-body problem is our concern, we should follow the link I made above between Feigl and C.I. Lewis, and ask: which notion of qualia is important for understanding their role in the mind-body problem? When Frank Jackson said that he is a qualia freak, and went on to argue that the existence of qualia presents a problem for physicalism, what did he mean by qualia? We can address this question by asking what qualia would have to be for qualia-based objections to physicalism to succeed. 51 For the main role which qualia play in contemporary debate is as the source of the central objection to physicalism: that physicalism cannot account for qualia. An argument against physicalism which simply appealed to the existence of certain mental states (e.g. sensations) and then claimed that physicalism could not account for these states is not an argument which should trouble physicalism. As David Lewis has put it, in discussing the sensation of pain, Pain is a feeling. To have pain and to feel pain are one and the same. For a state to be pain and for it to feel painful are likewise one and the same. A theory of what it is for a state to be pain is inescapably a theory of what it is like to be in that state, of how that state feels, of the phenomenal character of that state... Only if you believe on independent grounds that considerations of causal role and physical realisation have no bearing on whether a state is pain should you say that they have no bearing on how that state feels Frank Jackson, Epiphenomenal qualia Philosophical Quarterly , pp Mad pain and martian pain Philosophical Papers volume I (Oxford: Oxford University Press 1983). 22

23 The mere assertion that a physicalist cannot identify a supposed qualitative state with a brain state is not an objection, but a straightforward denial of physicalism. To put the point in this way is to beg the question. Jackson s well-known knowledge argument, however, does not beg the question in this way against physicalism. The question it raises (as Lewis goes on to note in a postscript to the paper just quoted) is Feigl s question: whether physicalism can account for our knowledge of qualia, and the answer it gives is that physicalism cannot do so, and that (surprisingly) there are therefore non-physical facts. The argument starts with a thought-experiment involving a person, Mary, who is virtually omniscient about colour and colour vision, who has lived all her life in a black-andwhite room, and has done all her learning there. When she leaves her room and sees something red for the first time, she comes to know something new: what it is like to see red. Let s call this knowledge of the qualia of red. Jackson argues that if you cannot know about qualia no matter what knowledge you have of the physical facts, then knowledge of qualia cannot be the same as knowledge of anything physical. So if facts are simply identified as the objects of knowledge, facts about qualia cannot be physical facts. I am not concerned here with whether the knowledge argument is successful, or with the physicalist responses to the argument. I am only concerned here with what has to be assumed about qualia if the argument is going to stand a chance of succeeding. For this will tell us what qualia are, insofar as qualia are intended to present a problem for physicalism. To begin with, we can ask what knowledge of qualia is supposed to be being contrasted with. We are told that Mary knows all the physical facts, so it might be thought that the contrast is between knowing everything physical, and knowing something non-physical. Hence the relevance of the conclusion to the truth of physicalism. But in fact, if the reasoning is sound, the conclusion is not just that physicalism is false, but something stronger. For it is not directly relevant that 23